Multiresponsive 4D Printable Hydrogels with Anti-Inflammatory Properties

Multiresponsive hydrogels are valuable as biomaterials due to their ability to respond to multiple biologically relevant stimuli, i.e., temperature, pH, or reactive oxygen species (ROS), which can be present simultaneously in the body. In this work, we synthesize triple-responsive hydrogels through UV light photopolymerization of selected monomer compositions that encompass thermoresponsive N-isopropylacrylamide (NIPAM), pH-responsive methacrylic acid (MAA), and a tailor-made ROS-responsive diacrylate thioether monomer (EG3SA). As a result, smart P[NIPAMx-co-MAAy-co-(EG3SA)z] hydrogels capable of being manufactured by digital light processing (DLP) 4D printing are obtained. The thermo-, pH-, and ROS-response of the hydrogels are studied by swelling tests and rheological measurements at different temperatures (25 and 37 °C), pHs (3, 5, 7.4, and 11), and in the absence or presence of ROS (H2O2). The hydrogels are employed as matrixes for the encapsulation of ketoprofen (KET), an anti-inflammatory drug that shows a tunable release, depending on the hydrogel composition and stimuli applied. The cytotoxicity properties of the hydrogels are tested in vitro with mouse embryonic fibroblasts (NIH 3T3) and RAW 264.7 murine macrophage (RAW) cells. Finally, the anti-inflammatory properties are assessed, and the results exhibit a ≈70% nitric oxide reduction up to base values of pro-inflammatory RAW cells, which highlights the anti-inflammatory capacity of P[NIPAM80-co-MAA15-co-(EG3SA)5] hydrogels, per se, without being necessary to encapsulate an anti-inflammatory drug within their network. It opens the route for the fabrication of customizable 4D printable scaffolds for the effective treatment of inflammatory pathologies.

M ultiresponsive hydrogels, also known as intelligent or smart hydrogels, can undergo controlled shape changes in response to more than one stimulus, which has attracted great attention in the biomedical field for drug delivery, 1 tissue engineering, 2,3 cancer therapy, 4 or biosensing. 5,6Their ability to change their properties upon response to biological (i.e., temperature, pH, enzyme activity, reactive oxygen species), and/or external stimuli, (i.e., light, electrical or magnetic field), 7−9 make them ideal candidates for 4D printing, a cutting-edge technology for manufacturing customizable dynamic materials combining 3D printing and stimuliresponsiveness. 10,11−16 Poly(Nisopropylacrylamide) (PNIPAM) is the most studied polymer to develop thermoresponsive hydrogels.PNIPAM displays a reversible volume phase transition through swelling at temperatures below the so-called lower critical solution temperature (LCST ∼ 32 °C) and shrinking above it. 3This swelling/shrinking process has been modulated through copolymerization with other monomers, which in turn confer responsiveness to other environmental stimuli such as pH. 12 For example, copolymers of PNIPAM and methacrylic acid (MAA) or acrylic acid (AA) can be deprotonated at high pHs, above their pK a , endowing hydrogels with pH-response in addition to temperature sensitivity. 17PMAA and PAA hydrogels have also been exploited for targeted drug release as they acted as drug protectors at acidic conditions in the stomach to be later released at higher pH ∼ 8 in the gastrointestinal tract. 18−21 More recently, the use of reactive oxygen species (ROS), oxidant species present in the human body, has drawn attention as possible stimuli for responsive hydrogels.ROS effect can vary from beneficial cell survival to nondesirable oxidative stress when they are overproduced, thus causing inflammation, cancer, and age-related diseases. 22,23−27 Very recently, we developed ROSresponsive photopolymerizable thioether-based hydrogels through the synthesis of aqueous soluble redox monomers from oligomers of ethylene glycol sulfur diacrylate (EG 3 SA).The resulting hydrogels were used as 5-Fluorouracil carriers to inhibit the growth of melanoma cancer cells. 28onsidering that overproduction of ROS in tumor/inflamed areas is generally linked to pH changes becoming slightly acidic (pH 5.4−7.1), 29,30the development of intelligent hydrogels that respond simultaneously to ROS, acidic pH, and body temperature is an interesting approach to modulate the simultaneous multistimulation in complex biological environments.Here, we have synthesized multiresponsive hydrogels through UV-light photopolymerization of a selected monomers mixture consisting of thermoresponsive NIPAM, pH-responsive MAA, and a tailor-made ROS-responsive EG 3 SA monomer.P[NIPAM x -co-MAA y -co-(EG 3 SA) z ] hydrogels' response to external stimuli (temperature, pH, and ROS) and controlled-release properties are investigated.Their additive manufacturing through digital light processing (DLP) 4D printing is also performed to obtain customizable hydrogels.Finally, we show that the hydrogels display anti-inflammatory properties.
The chemical characterization of the hydrogels was performed by infrared spectroscopy (Figure S1).The peak at 1720 cm −1 is attributed to C�O vibrations of the acid carbonyl groups of MAA and the acrylate groups of EG 3 SA.The peaks at 1650, 1540, and 1130 cm −1 are assigned to C� O, N�H, and C�N stretching of amide groups present in NIPAM.The peak at around 1455 cm −1 is attributed to C�H bending in the −(CH 3 ) 2 and −CH 2 groups of NIPAM and MAA, 31 and the peaks at 690 and 715 cm −1 are the signatures of symmetric and asymmetric dimethyl sulfide bonds, respectively.After H 2 O 2 treatment, thioether groups of EG 3 SA are oxidized into sulfoxides and/or sulfones, as corroborated by the appearance of two peaks at 1020 and 1320 cm −1 corresponding to the stretching of the double bond S�O in sulfoxides and O�S�O in sulfones, respectively. 28he response of the hydrogels to different stimuli, ROS, temperature, and pH, was tested (Figure 2a).different pHs (Figure 2d): P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] hydrogels experienced a huge swelling of ≈465 wt % at pH 7.4 due to the oxidation of the thioether groups present in the EG 3 SA domains, which led to more hydrophilic hydrogels.This effect was less pronounced in hydrogels with a larger content of EG 3 SA due to their higher cross-linking density.All hydrogels exhibited the highest swelling properties in H 2 O 2 at pH 11, as all copolymer components are in the most hydrophilic state, holding the largest quantity of water, ≈1320 wt % for P[NIPAM 80 -co-MAA 15   shrinking, while the pH increase to 11 increased the release (0.33 mg/mL).In the presence of H 2 O 2 at pH 7.4 and 37 °C, the release of KET increased (0.27 mg/mL) due to the oxidation of the thioether groups of EG 3 SA becoming more hydrophilic, while the decrease of pH slightly reduced the KET released (0.25 mg/mL).This hydrophilic effect is more evident over time as the release of KET increased after 72 h.swelling capacity.Therefore, all hydrogels were thermo-, pH-, and ROS-responsive leading to tunable KET release profiles.
The additive manufacturing by DLP 3D printing was initially studied by photorheology (Figure 4a).Before irradiation (0− 60 s), the loss modulus (G′′) was higher than the storage modulus (G′), pointing out the liquid-like state of the copolymer inks.After 60 s, the UV light was switched and the photopolymerization process started reaching a solid-like state (G′ > G′′) in a few seconds.The photopolymerization time decreased with an increase in the EG 3 SA concentration from 20 s for P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] to 12 and 10 s for P[NIPAM 70 -co-MAA 15 -co-(EG 3 SA) 15 ] and P-[NIPAM 40 -co-MAA 20 -co-(EG 3 SA) 40 ], respectively.As no significant differences were observed between hydrogels with 5% and 15% mol EG 3 SA, only hydrogels with the lowest and highest percentage of this monomer, 5% and 40% mol, were studied in further experiments.The copolymers with 5% and 40% mol EG 3 SA were successfully processed by DLP 3D printing to fabricate customized multihollow scaffolds (Figure 4b).Thanks to their stimuli-responsive properties, they became 4D-printable hydrogels.The printing resolution increased with the percentage of EG 3 SA monomer within the copolymers, but the hydrogels were more brittle.The mechanical properties in the presence of different stimuli were characterized by rheology.In all cases, G′ was higher than G′′, corroborating the hydrogel formation (Figure 4c).Three different conditions were tested to characterize the hydrogels based on the most representative biological conditions: (i) Physiological mimicking conditions (PBS, pH 7.4, 37 °C): G′ increased with NIPAM percentage within the hydrogels, from ≈2 × 10 4 Pa for P[NIPAM 40 -co-MAA 20 -co-(EG 3 SA) 40 ] (Figure S3) to ≈6 × 10 4 Pa for P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] due to their higher contraction at physiological temperature making them less flexible.(ii) Oxidation in the presence of ROS (H 2 O 2 ), which are present in inflammatory diseases, 32 at pH 7.4 and 37 °C (Figure 4d).The mechanical properties were highly influenced by the thermoresponse of NIPAM and ROS-response of EG 3 SA.G′ increased up to ≈1 × 10 5 Pa in P[NIPAM 40 -co-MAA 20 -co-(EG 3 SA) 40 ] hydrogels due to the higher percentage of EG 3 SA that led to the formation of sulfoxides and sulfones, thus, allowing them to hold a high quantity of water and making them more brittle.(iii) Oxidation (H 2 O 2 ) at pH 5 and 37 °C (Figure 4e).No significant differences were observed in G′ of P[NIPAM 40 -co-MAA 20 -co-(EG 3 SA) 40 ] hydrogels between pH 7.4 and pH 5, probably because they reached the maximum swelling capacity before breaking.G′ of P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] hydrogels decreased to ≈3 × 10 4 Pa because of the less amount of cross-linker EG 3 SA and the higher elasticity of the oxidized chains.
The cytotoxicity of nonloaded and KET-loaded P[NIPAM xco-MAA y -co-(EG 3 SA) z ] hydrogels was tested in vitro with mouse embryonic fibroblasts (NIH 3T3) and RAW 264.7 murine macrophage (RAW) cells.These cell lines were selected as representative models involved in inflammatory processes during tissue repair, 33−36 in which macrophages modulate inflammation and fibroblasts lay down a new extracellular matrix.P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] hydrogels were not cytotoxic, showing NIH 3T3 and RAW cell viabilities higher than 90% (Figure 5a,b).However, P-[NIPAM 40 -co-MAA 20 -co-(EG 3 SA) 40 ] hydrogels reduced the viability of NIH 3T3 (≈ 85%) and RAW (≈ 65%) cells and were discarded for the next experiments.It was observed that the presence of the EG 3 SA within P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] hydrogels favored the NIH 3T3 cell adhesion in comparison with P[NIPAM 90 -co-MAA 10 ] hydrogels (Figure 5c,d).NIH 3T3 cell morphology was visualized by staining cell nuclei with Hoechst (blue staining) and cytoskeleton (F-actin fibers) with phalloidin-rhodamine (orange staining).On P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] hydrogels, NIH 3T3 cells showed an elongated morphology and cell spreading, which was not observed in P[NIPAM 90 -co-MAA 10 ] hydrogels where fibroblasts exhibited a round morphology forming clusters.It is known that cell adhesion is influenced by chemical groups present on the surface of the hydrogels 37 and specifically enhanced by sulfonic groups, 38 which induce a reorganization of the actin cytoskeleton of fibroblasts. 39,40hus, the thioether groups present on the EG 3 SA domains favored the cell adhesion on P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] hydrogels.The anti-inflammatory properties of nonloaded and KET-loaded P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] hydrogels were tested in contact with RAW cells, which can polarize to their pro-inflammatory phenotype (M1) when they are activated by lipopolysaccharide (LPS) and start to overproduce nitric oxide (NO).The anti-inflammatory capacity of the hydrogels was determined by measuring the NO production of LPS-activated RAW cells (LPS-RAW) seeded on the hydrogels for 24 h (Figure 5e) in comparison with LPS-RAW cells seeded on the well plate (positive control, inflammatory conditions untreated -ICU) and non-LPSactivated RAW cells seeded on the hydrogels (negative control, noninflammatory conditions -NIC).The NO released by LPS-RAW cells seeded on top of P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] hydrogels decreased up to 29.9 ± 2.6% after 24 h, reaching the basal value of non-LPS-activated RAW cells on the hydrogels (NIC = 32.3 ± 1.1%).Nonsignificant differences were detected in comparison with KET-loaded P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] hydrogels that also decreased the NO production (30.1 ± 4.4%) up to basal values.To quantify the anti-inflammatory capacity of P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] hydrogels, NO values were compared with those of LPS-RAW cells seeded on a plate and brought in contact with different ketoprofen concentrations (Figure S4).Results showed that NO released by LPS-RAW cells decreased up to basal values from 0.3 mg/mL KET approximately.Overall, these results proved the excellent anti-inflammatory capacity of P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] hydrogels, per se, without encapsulating an anti-inflammatory drug (Figure 5e).At the initial stage, RAW cells are seeded on top of P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] hydrogels.They are in a noninflammatory stage and NO production is at the basal value.In the second stage, the inflammatory process is induced by activation of RAW cells with LPS and they start to overproduce NO (a type of ROS).Then, in the last stage, the ROS produced by LPS-RAW cells are trapped by the P[NIPAM 80 -co-MAA 15 -co-(EG 3 SOA) 5 ] hydrogels in the oxidized EG 3 SA domains formed by sulfoxides and sulfones (EG 3 SOA), thus reducing the inflammatory process and the NO production by RAW cells up to basal values of noninflammatory conditions.This interesting achievement opens the route for the fabrication of 4D printable anti-inflammatory scaffolds in a customized manner with anti-inflammatory properties.
In conclusion, triple-responsive hydrogels were synthesized by photopolymerization of thermoresponsive NIPAM, pHresponsive MAA, and ROS-responsive EG 3 SA monomers.Thus, it also allowed their additive manufacturing by DLP leading to 4D printed shape-defined hydrogels.The swelling properties of P[NIPAM x -co-MAA y -co-(EG 3 SA) z ] hydrogels decreased with the temperature increase from 25 to 37 °C, due to the contraction of PNIPAM chains above their LCST, and with the pH decrease from 7.4 to 3, because of the protonation of the carboxylic groups of PMAA below their pK a .On the contrary, swelling increased with the pH increase from 7.4 to 11, due to the deprotonation of carboxylic groups, and with the presence of ROS (i.e., H 2 O 2 ) because of the oxidation of the thioether groups in EG 3 SA into sulfoxides and sulfones making them more hydrophilic.P[NIPAM x -co-MAA y -co-(EG 3 SA) z ] hydrogels were used as carriers for the controlled release of ketoprofen.The mechanical properties of the hydrogels were characterized by rheology showing a variation of the initial G′ values (≈10 4 Pa) that depended on the applied stimuli and could reach up to ≈10 5 Pa.Cell tests pointed out P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] hydrogels presented the optimal stimuli-responsive performance while being noncytotoxic, at the same time they possessed anti-inflammatory properties per se.

Figure 1 .
Figure 1.(a) Chemical route employed for the synthesis of P[NIPAM x -co-MAA y -co-(EG 3 SA) z ] hydrogels by UV light at 365 nm for 3−5 min and using Darocur as a photoinitiator.(b) Pictures of the synthesized hydrogels with different monomer ratios.
-co-(EG 3 SA) 5 ], ≈615 wt % for P[NIPAM 70 -co-MAA 15 -co-(EG 3 SA) 15 ], and ≈265 wt % for P[NIPAM 40 -co-MAA 20 -co-(EG 3 SA) 40 ].(iv) Triple ROS-, thermo-, and pH-response under oxidative conditions (9 mM H 2 O 2 ) at 37 °C and different pHs, where all monomers are involved in the stimuli-responsive properties (Figure 2e): The least cross-linked P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] hydrogels presented the highest swelling at all pHs (Figures 1f,g and S2).At pH 7.4, P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5] hydrogels showed the lowest swelling (≈93 wt %), as they contained the highest percentage of PNIPAM.The swelling decreased with the pH up to ≈8 wt % at pH 3 due to the combination of the shrinking behavior of PNIPAM and PMAA that camouflaged the hydrophilic oxidation properties of PEG 3 SA.At alkaline pH 11, the swelling increased exponentially due to the deprotonation of COO − groups of PMAA together with the more hydrophilic oxidized PEG 3 SA, which led to the hydrogel's breaking, probably due to the high pressure produced by the water, which broke their network and made them difficult to handle.P[NIPAM n -co-MAA m -co-(EG 3 SA) x ] hydrogels were tested as scaffolds to encapsulate an anti-inflammatory drug, ketoprofen (KET).The KET release under different stimuli (temperature, pH, ROS; Figure3a,b) was correlated with the hydrogels' swelling behavior (Figure2).P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] and P[NIPAM 70 -co-MAA 15 -co-(EG 3 SA) 15 ] hydrogels showed a much higher capability of releasing KET than P[NIPAM 40 -co-MAA 20 -co-(EG 3 SA) 40 ], in agreement with swelling tests.P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] hydrogels released 0.26 mg/mL of KET after 24 h in PBS at pH 7.4 and 25 °C.The release decreased up to 0.21 mg/mL at 37 °C due to the NIPAM-induced shrinking trapping a higher part of KET molecules inside.The decrease of pH to 5, at 37 °C, induced a slight reduction of the KET released up to 0.19 mg/mL because of the MAA-induced
P[NIPAM 70 -co-MAA 15 -co-(EG 3 SA) 15 ] hydrogels showed a similar behavior.The same trends were observed in the case of P[NIPAM 40 -co-MAA 20 -co-(EG 3 SA) 40 ] hydrogels, although the concentration of KET released was much lower due to their higher cross-linking degree and consequently lower

Figure 3 .
Figure 3. (a) Schematic representation of the ketoprofen (KET) release from P[NIPAM x -co-MAA y -co-(EG 3 SA) z ] hydrogels under different conditions of temperature, pH, and ROS.(b) Release of KET from P[NIPAM x -co-MAA y -co-(EG 3 SA) z ] hydrogels under nonoxidative conditions in PBS, and oxidative conditions in the presence of 9 mM H 2 O 2 , at 25 or 37 °C, and pH 5, 7.4, or 11 for 24 and 72 h.Diagrams include the mean and standard deviation (n = 3) and the ANOVA results.Different letters indicate statistically significant differences at a significance level of p < 0.05 using Tukey's test.Bars with no common letters are significantly different (p < 0.05).

Figure 5 .
Figure 5.In vitro cytotoxicity tests of nonloaded and KET-loaded P[NIPAM x -co-MAA y -co-(EG 3 SA) z ] hydrogels in contact with (a) NIH 3T3 and (b) RAW cells for 24 and 48 h.(c) NIH 3T3 cell adhesion on (c) P[NIPAM 90 -co-MAA 10 ] and (d) P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] hydrogels.(e) Nitric oxide (NO) released by LPS-RAW cells seeded on the plate (ICU), non-LPS-RAW cells seeded on the plate (NIC), and LPS-RAW cells seeded on the hydrogels.Diagrams (a), (b), and (e) include the mean and standard deviation (n = 3) and the ANOVA results at significance levels of *p < 0.05 and ***p < 0.001 using Tukey's test (ns means nonstatistically significant differences).(f) Schematic representation of the inflammatory process induced by the activation of RAW cells, growth on top of the hydrogels, with LPS leading to ROS production, and the antiinflammatory properties of P[NIPAM 80 -co-MAA 15 -co-(EG 3 SA) 5 ] hydrogels, which trapped ROS due to the oxidative capacity of the (EG 3 SA) block.

Materials and methods, 1 H
NMR spectra of EG 3 SA monomer in CDCl 3 , FTIR spectra of P[NIPAM 70 -co-MAA 15 -co-(EG 3 SA) 15 ] hydrogels at different conditions, representative pictures of P[NIPAM x -co-MAA y -co-(EG 3 SA) z ] hydrogels under different conditions and swelling comparison, rheological properties of P-[NIPAM x -co-MAA y -co-(EG 3 SA) z ] hydrogels in PBS at pH 7.4 and 25 °C, and nitric oxide (NO) released by RAW cells in the presence of different ketoprofen concentrations (PDF) ■ AUTHOR INFORMATION Corresponding Authors Sergio E. Moya − Center for Cooperative Research in Biomaterials (CIC biomaGUNE), 20014 Donostia-San